Tim wrote: > > It wasn't that chicken-and-egg situation that I was referring to, but > the long standing issue that the self check has been bad for a very long > time. If *something* *else* can manage to not read past the end of the > disc, why can't it? Ok, I misunderstood what you were referring to. Yes, that should be checkable, but that's not the best check. It would be better to put MD5 (or SHA1) sums inside the file system, and check the files, rather than the image, anyway. That obviates the entire problem, and provides a better check. [...] > You'd have a routine that asked for a checksum to be inputed, from some > other source, and the disc to be checked against it. Not necessary. Check the files from *inside* the file system, not outside. All files get an MD5 or SHA1 check performed on them, except for the one file which has all the sums. That one simply gets a CRC (since it won't be very big). There are already means to embed such a CRC in the program which does the checks. It can then check itself first, via CRC, then perform the MD5 or SHA1 checks on all the other files. That is, for self-checks. The program first checks itself via CRC, then checks the files on the image, then asks for all the other discs. It has the MD5 (SHA1) sums for all the others. For those reads, you would want to mount the disc, and then check the files inside the file system. For checks of basic downloads, and burt images (not self-checks) you'd want to be able to do what you suggest. Mike -- p="p=%c%s%c;main(){printf(p,34,p,34);}";main(){printf(p,34,p,34);} Oppose globalization and One World Governments like the UN. This message made from 100% recycled bits. You have found the bank of Larn. I speak only for myself, and I am unanimous in that! -- users mailing list users@xxxxxxxxxxxxxxxxxxxxxxx To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines